The intention is to provide this as part of PrologHub, which will be changing so that the blog is only one part of it. PrologHub is aiming to be a "goto" place for people learning Prolog, so it will provide this service as well as cheat-sheets, reading recommendations, topic specific tutorials and learning projects.
By tracking what is being learnt it is our hope that Learn Prolog Now! can be integrated with the other services such that if we notice someone taking a project doesn't know something taught in the book we can point them towards the relevant section.
Requires Logtalk (developing with 3.28.0). This can be installed with the query:
?- pack_install(logtalk).
The web-server is in www/server.pl. In the www directory I tend to
run swipl -g serve -s server.pl
- www contains source and server for website
- www/book contains all the Logtalk Objects that describe the content
- www/html_components contains html components... surprise!
We're using termarized HTML, currently copy-pasted and then transformed. The current websites don't include all the formatting present in the Latex/Printed Book, so we're adding this back in. It's probably possible to write a parser that'll take the Latex and spit out termarized HTML, but it might prove to take longer.
Termarized HTML is being used to create components, such as a code query or navigation page. Where possible this should be improved upon.
We're strictly using Bootstrap for a professional style and to integrate into PrologHub. For these reason we're using the same colour scheme.
We're using JQuery, as per Bootstraps requirements. We've got prism.js
providing client-side syntax highlighting too. Finally there's some
homespun JavaScript to interact with Tau-Prolog. To avoid writing quiz
logic in JavaScript, these are also handed off to Tau-Prolog.
localStorage is being used to store the nav-sidebar state between page
loads. This can be cleared with localStorage.clear() should it need to
be refreshed.
Tau-Prolog is providing all the Prolog interactivity. Changing the
code-block structure for \code_query will likely break the JavaScript
due to how it navigates the DOM.
There are many types of these... See book/quizzes.lgt for the
implemented ones. Each quiz has a required_script, which is the
JavaScript to be loaded for the interactivity. These are stored in
static/js/quizzes/
It would be nice to provide an in-browser editor for the user. Thought is required as to whether Tau-Prolog is sufficient for this or if we should integrate SWISH.
Besides logging in, we need to integrate user data into the
ontology. When a user requests a resource this will add a triple, when
they successfully complete an assessment this will also add a triple.
Thus we can track what they've been taught and learnt.
library(persistency) will be good for this. It also means we can allow
a user to download all their own data (and static ontology) in Prolog
for their own amusement. User account data can be stored in "FOAF"
inspired triples.
Not yet deleted as it contains data not yet transferred.
archive contains old stuff (just for the record)
web_site contains sources of OLD website
prolog example prolog databases
text contains latex sources and produced ps and pdf files
scripts contains shell/perl scripts for conversion and pdf production
www contains sources of current website
scripts/_generate_html converts html from latex sources
scripts/_run produces lpn.ps and lpn.pdf
The sources are ISO-Latin-1 files. They do not compile after conversion to UTF-8. The scripts must be executed under non-UTF-8 locale.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent